Hope in a Hopeless Place: Validating the Kessler-6 Scale of Psychological Distress in the Incarceration Context

• In Event: 20820 - Section on Medical Sociology Roundtables
  In Refereed Roundtable (90 minute): Table 6. Mental Health

Abstract

Extensive research has focused on the mental health implications of incarceration, including the psychological outcomes of prisoners and methodological issues behind identifying causal effects. Yet an important issue remains overlooked: the conventional mental health batteries, such as Kessler-6 scales of psychological distress, were validated in civilian populations but never among prisoners. Given the profound roles of the total institution in shaping prisoners’ values, daily routines, and life prospects, there are strong sociological grounds to believe that certain measures (e.g., worthlessness, hopelessness) hold different meanings for inmates, which are potentially conflated with psychological well-being. Drawing on harmonized data from the 2016 National Health Interview Survey and 2016 Survey of Prison Inmates, we carry out the first analysis that validates Kessler-6 items in the incarceration context, among the overall inmate sample and stratified inmate samples by the length of remaining sentence – an important temporal factor that shapes prisoners’ well-being. We first provide a sociological framework on the substantive origins of K6 items’ measurement biases in the incarceration context. Then, analyzing data with psychometric methods, we found that: 1) Several K6 items – restless, worthless, and hopeless exhibited significant measurement biases among inmates; 2) their biases are primarily concentrated among inmates with shorter remaining sentences, except for hopeless, which was severely biased for understanding inmates with longer terms left behind bars; 3) purifying the scales contributed to uncovering a curvilinear relationship between psychological distress and length of remaining sentence, a pattern not observed using original scales. We conclude by discussing how the K6, in tandem with an appreciation of its biases, can be used to provide a more accurate picture of individual well-being for the currently incarcerated population.

Authors

• Yezhen Li, University of Southern California

• Jason Schnittker, University of Pennsylvania